Authors :
Presenting Author: Negar Noorizadeh, PhD – Department of Pediatrics, Division of Pediatric Neurology, University of Tennessee Health Science Center, Memphis, TN, USA; Neuroscience Institute, Le Bonheur Children’s Hospital, Memphis, TN, USA;
Seth B. Crum, MS – College of Medicine, University of Tennessee Health Science Center, Memphis, TN, USA;; Talitha Boardman, BSc – Department of Pediatrics, Division of Pediatric Neurology, University of Tennessee Health Science Center, Memphis, TN, USA; Neuroscience Institute, Le Bonheur Children’s Hospital, Memphis, TN, USA;; Savannah Gibbs, Ms – Neuroscience Institute, Le Bonheur Children’s Hospital, Memphis, TN, USA;; James W. Wheless, MD – Professor, Department of Pediatrics, Division of Pediatric Neurology, University of Tennessee Health Science Center, Memphis, TN, USA; Neuroscience Institute, Le Bonheur Children’s Hospital, Memphis, TN, USA;; Shalini Narayana, PhD – Professor, Department of Pediatrics, Division of Pediatric Neurology, University of Tennessee Health Science Center, Memphis, TN, USA; Neuroscience Institute, Le Bonheur Children’s Hospital, Memphis, TN, USA; Department of Anatomy and Neurobiology, University of Tennessee Health Science Center, Memphis, TN, USA
Rationale:
Functional brain mapping with Transcranial Magnetic Stimulation (TMS) has emerged as a reliable and effective non-invasive method for localizing eloquent cortices including speech and language areas. TMS induces language errors by temporarily disrupting specific brain activity, particularly in the temporal and frontal regions. Although the TMS-derived speech and language maps are obtained for use at an individual level, a speech and language atlas can be derived by combining statistical models and TMS data to estimate the location and variability of corresponding cortices in a clinical cohort. In this study, we have created atlases of critical speech and language regions in patients with epilepsy or brain tumor based on age and TMS-induced error type.
Methods:
In a retrospective chart review, we identified 316 patients who had undergone successful TMS language mapping as part of Phase 1 pre-surgical evaluation (Table 1). TMS-induced speech errors were classified into three categories: speech arrest, where TMS caused a complete inability to produce any verbal response; semantic error, wherein a word with a similar meaning or association was used instead of the target word; and performance error that included distortions, slurring, stuttering, imprecise articulation, or delayed responses compared to the baseline. To create a speech and language atlas, the data were first transformed to the space of an age-specific standard MNI atlas. A coordinate-based activation likelihood estimation (ALE) meta-analysis was then performed using GingerALE (Brainmap.org) for each error type in two age groups (children £ 18 years and adults > 18 years). Created atlases provided insight into the spatial distribution and prevalence of speech errors induced by TMS.
Results:
The distribution of speech and language areas where the three types of errors were observed in the pediatric and adult patients is shown in Figure 1
. The findings revealed that all three types of errors were observed in both hemispheres, specifically within the frontal and temporal lobes. This pattern was consistent across both children and adults. However, notable differences were identified between the two age groups. In children, speech arrest was more prominently observed in the left hemisphere compared to adults. Additionally, brain areas corresponding to semantic processing was not localized in the right hemisphere of children.Conclusions:
This study is the first to create an atlas of speech and language areas based on TMS-induced errors across different ages in a clinical cohort with epilepsy or brain tumor. Our findings revealed critical speech and language areas in the temporal and frontal lobes of both hemispheres in both adult and pediatric patients. These innovative atlases can be integrated as a plugin in widely used imaging software, offering an alternative method for education, evaluating language (re)organization. They can guide treatment planning to enhance patient outcomes within individuals and centers without TMS access by accurately identifying speech and language areas overlapping with epileptic or tumor regions before surgery.
Funding: N/A